Tumor necrosis factor (TNF) is a monokine potentially useful as a therapy for neoplastic disorders which also functions as a central mediator of the cachectic state and lethal effects of endotoxin in mice. Interaction of TNF with endothelium could provide a common denominator for the elicitation of responses in multiple organ systems, characteristic of the host response to inflammatory and neoplastic stimuli. This proposal is motivated by the hypothesis that endothelium is a primary target mediating the biologic effects of TNF and that TNF-induced perturbation of endothelium is critical for tumor necrosis. Preliminary data using cultured endothelium indicate that TNF does modulate endothelial cell coagulant properties, Interleukin-1 generation, proliferation and integrity of the monolayer providing powerful mechanisms for interrupting the blood supply to neoplastic lesions and inducing toxic effects in normal tissues. We seek to understand mechanisms underlying these changes in endothelial cell physiology induced by TNF and correlate them with in vivo findings following infusion of TNF into normal and tumor-bearing animals. A potentially important link between endothelial cell coagulation factor receptors and TNF-induced modulation of endothelial function will be explored: occupancy of endothelial cell activated protein C/protein S (APC/PS) sites decreases TNF binding to its receptor on endothelium and prevents several changes induced by TNF in cell surface coagulant properties. The ability of APC/PS to attenuate the endothelial cell response to TNF will be studied in the context of endothelial cell coagulant, growth and morphologic properties. Mechanisms involved in modulation of the cellular TNF response by APC/PS will be examined at the level of TNF-receptor interaction and second messenger pathways including cyclo-oxygenase metabolites and cytosolic calcium flux. The capacity of APC/PS to regulate the TNF response of two types of physiologically altered endothelium, Gamma-interferon-treated cells and preconfluent endothelium will be contrasted with confluent endothelial cells. If the contribution of APC/PS is different in these situations, this would suggest that occupancy of endothelial cell APC/PS sites may provide a mechanism for targeting TNF effects to the tumor vascular beds. This hypothesis will be tested in infusion studies in which TNF and APC/PS will be given to normal and tumor-bearing mice. Results of these studies should contribute to the understanding of mechanisms underlying TNF-endothelial cell interaction, thereby potentially optimizing TNF action in the tumor vascular bed, while minimizing effects in the surrounding tissues.